INDEX

Explanations

code that involves class definitions and programming constructs

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 فريبيس

-0.79

ształ

-0.70

 arşivlendi

-0.58

bnf

-0.54

resave

-0.54

elif

-0.53

ówczas

-0.51

COIN

-0.50

setVerticalGroup

-0.50

kste

-0.50

POSITIVE LOGITS

expandindo

0.85

Personensuche

0.61

providedIn

0.60

public

0.59

 trasparente

0.58

SourceChecksum

0.58

private

0.58

дивиду

0.58

 Protección

0.57

 privata

0.56

Activations Density 0.019%

code that involves class definitions and programming constructs

No Comments

No Known Activations